Editing drafts in the queue in ISO-8859-2

Editing drafts in the queue in ISO-8859-2

[Pavel Janik ml.]

Hello,

I have

   (set-language-environment "czech")
   (set-selection-coding-system 'iso-8859-2)
   (setq process-coding-system-alist '((".*" . iso-8859-2)))

But when I want to edit an draft in the outgoing queue (D e) with agent, the
mail which is of course in ISO 8859-2 is displayed with those nasty \x948 like
chars instead of ISO 8859-2 characters.

How can I avoid it? Is this repaired in CVS version?
-- 
Pavel Janík ml.
Pavel.Janik@inet.cz

Re: Editing drafts in the queue in ISO-8859-2

[Lars Magne Ingebrigtsen]

Pavel.Janik@inet.cz (Pavel Janik ml.) writes:

>    (set-language-environment "czech")
>    (set-selection-coding-system 'iso-8859-2)
>    (setq process-coding-system-alist '((".*" . iso-8859-2)))
> 
> But when I want to edit an draft in the outgoing queue (D e) with
> agent, the mail which is of course in ISO 8859-2 is displayed with
> those nasty \x948 like chars instead of ISO 8859-2 characters.

Hm.  I wouldn't have been surprised if the charset was wrong, but
\x948 would lead me to believe that `prin' or its friends have been
involved at some point, which `D e' shouldn't be doing.

Or perhaps you didn't mean a "x" after the backslash?

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Editing drafts in the queue in ISO-8859-2

[Pavel Janik ml.]

   From: Lars Magne Ingebrigtsen <larsi@gnus.org>
   Date: 20 Apr 2000 02:16:19 +0200

Hi,

   > Hm.  I wouldn't have been surprised if the charset was wrong, but
   > \x948 would lead me to believe that `prin' or its friends have been
   > involved at some point, which `D e' shouldn't be doing.
   > 
   > Or perhaps you didn't mean a "x" after the backslash?

no. I dig deeply and here are the results.

1. I composed test message with ěščřžýáíé (ISO-8859-2) letters in the body.
2. C-c C-c and the message is stored in the queue.
3. I went to the queue group and saw the message in the summary. The message
itself looks OK in the message buffer. I tried to D e on it. I received
\357\251 in the body. The numbers are random (i.e. not exactly the same I've
got, but you got the point :-). When I press C-x k RET here, the message goes
into drafts folder from the queue.

BAD.

1. I composed another test message with ěščřžýáíé (ISO-8859-2) letters in the
   body and sent it to queue.
2. I tried to press `e' in the summary. Everything is ok and the message _is_
   readable with correct charset. I can add another ISO 8859-2 characters in
   it and they are showed ok. Then I pressed C-c C-c to end the edits. The
   message buffer is then switched into \357\251 chars like above. The I can
   show another message and came back to my test message again and it is
   showed in the message buffer correctly. And it is sent correctly.

Hmm, I accidentally pressed D e in the message buffer in the queue group and
got an error message. Hm, D is gnus-article-read-summary-keys in the message
buffer but when I'm in the message buffer in the queue group, Emacs is waiting
for another key and shows D- in the minibuffer... Maybe this will help you to
debug it.

Backtrace:

Signaling: (wrong-type-argument commandp 1)
  call-interactively(1)
  gnus-article-read-summary-keys(nil)
* call-interactively(gnus-article-read-summary-keys)
  recursive-edit()
  byte-code("Ć\b!ˆ \x11Č ˆ !ˆpĘË\x1a^[\x1cĚ ˆ+ebˆ  ˆ`ÎĎ!ˆ yˆ`|ˆ  ˆ\r@Ň>ƒN
  debug(error (wrong-type-argument commandp 1))
  call-interactively(1)
  gnus-article-read-summary-keys(nil)
* call-interactively(gnus-article-read-summary-keys)
  recursive-edit()
  byte-code("Ć\b!ˆ \x11Č ˆ !ˆpĘË\x1a^[\x1cĚ ˆ+ebˆ  ˆ`ÎĎ!ˆ yˆ`|ˆ  ˆ\r@Ň>ƒN
  debug(error (wrong-type-argument commandp 1))
  call-interactively(1)
  gnus-article-read-summary-keys(nil)
* call-interactively(gnus-article-read-summary-keys)

GNU Emacs 20.6/GNUS 5.8.4
-- 
Pavel Janík ml.
Pavel.Janik@inet.cz

Re: Editing drafts in the queue in ISO-8859-2

[Shenghuo ZHU]

>>>>> Pavel Janik ml <Pavel.Janik@inet.cz> writes:

 >    From: Lars Magne Ingebrigtsen <larsi@gnus.org>
 >    Date: 20 Apr 2000 02:16:19 +0200

 > Hi,

>> Hm.  I wouldn't have been surprised if the charset was wrong, but
>> \x948 would lead me to believe that `prin' or its friends have been
>> involved at some point, which `D e' shouldn't be doing.
>> 
>> Or perhaps you didn't mean a "x" after the backslash?

 > no. I dig deeply and here are the results.

 > 1. I composed test message with ěščřžýáíé (ISO-8859-2) letters in the body.
 > 2. C-c C-c and the message is stored in the queue.
 > 3. I went to the queue group and saw the message in the summary. The message
 > itself looks OK in the message buffer. I tried to D e on it. I received
 > \357\251 in the body. The numbers are random (i.e. not exactly the same I've
 > got, but you got the point :-). When I press C-x k RET here, the message goes
 > into drafts folder from the queue.

 > BAD.

It is because the message from nndraft:queue is not decoded.  Now,
this bug is fixed in 5.8.5. 

 > 1. I composed another test message with ěščřžýáíé (ISO-8859-2) letters in the
 >    body and sent it to queue.
 > 2. I tried to press `e' in the summary. Everything is ok and the message _is_
 >    readable with correct charset. I can add another ISO 8859-2 characters in
 >    it and they are showed ok. Then I pressed C-c C-c to end the edits. The
 >    message buffer is then switched into \357\251 chars like above. The I can
 >    show another message and came back to my test message again and it is
 >    showed in the message buffer correctly. And it is sent correctly.

 > Hmm, I accidentally pressed D e in the message buffer in the queue group and
 > got an error message. Hm, D is gnus-article-read-summary-keys in the message
 > buffer but when I'm in the message buffer in the queue group, Emacs is waiting
 > for another key and shows D- in the minibuffer... Maybe this will help you to
 > debug it.

 > Backtrace:

 > Signaling: (wrong-type-argument commandp 1)
 >   call-interactively(1)
 >   gnus-article-read-summary-keys(nil)
 > * call-interactively(gnus-article-read-summary-keys)

[...]

lookup-key returns 1. Weird. This one is not fixed.

-- 
Shenghuo

Re: Editing drafts in the queue in ISO-8859-2

[Lars Magne Ingebrigtsen]

Pavel.Janik@inet.cz (Pavel Janik ml.) writes:

> Hmm, I accidentally pressed D e in the message buffer in the queue
> group and got an error message. Hm, D is
> gnus-article-read-summary-keys in the message buffer but when I'm in
> the message buffer in the queue group, Emacs is waiting for another
> key and shows D- in the minibuffer... Maybe this will help you to
> debug it.

Do you mean `D e' in the article buffer?

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Editing drafts in the queue in ISO-8859-2

[Pavel Janik ml.]

   From: Lars Magne Ingebrigtsen <larsi@gnus.org>
   Date: 21 Apr 2000 13:56:59 +0200

   > > Hmm, I accidentally pressed D e in the message buffer in the queue
   > > group and got an error message. Hm, D is
   > > gnus-article-read-summary-keys in the message buffer but when I'm in
   > > the message buffer in the queue group, Emacs is waiting for another
   > > key and shows D- in the minibuffer... Maybe this will help you to
   > > debug it.
   > 
   > Do you mean `D e' in the article buffer?

yes.
-- 
Pavel Janík ml.
Pavel.Janik@inet.cz

Re: Editing drafts in the queue in ISO-8859-2

[Lars Magne Ingebrigtsen]

Pavel.Janik@inet.cz (Pavel Janik ml.) writes:

>    > Do you mean `D e' in the article buffer?
> 
> yes.

It doesn't look like the summary key reading thing isn't working as it
should with multi-key keystrokes...

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Editing drafts in the queue in ISO-8859-2

[Shenghuo ZHU]

>>>>> "LMI" == Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

LMI> Pavel.Janik@inet.cz (Pavel Janik ml.) writes:

>> > Do you mean `D e' in the article buffer?
>> 
>> yes.

LMI> It doesn't look like the summary key reading thing isn't working
LMI> as it should with multi-key keystrokes...

It is because `D e' is a minor mode keystroke. I tried to fixed it,
but I saw comments like "We disable the pick minor mode commands".
BTW, in Emacs 20.6, (lookup-key (current-local-map) "De") returns 1
(not null), which results the back-trace.

-- 
Shenghuo

Re: Editing drafts in the queue in ISO-8859-2

[Lars Magne Ingebrigtsen]

Shenghuo ZHU <zsh@cs.rochester.edu> writes:

> BTW, in Emacs 20.6, (lookup-key (current-local-map) "De") returns 1
> (not null), which results the back-trace.

Yup.  Fix in Gnus v5.8.5.

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen